System and method for efficient network reconfiguration in fat-trees

Abstract

Systems and methods are provided for supporting efficient reconfiguration of an interconnection network having a pre-existing routing. An exemplary method can provide a plurality of switches, a plurality of end nodes, and one or more subnet managers, including a master subnet manager. The method can calculate, via the master subnet manager, a first set of one or more leaf-switch to leaf-switch multipaths. The method can store this first set of one or more leaf-switch to leaf-switch multipaths at a metabase. The method can detect a reconfiguration triggering event, and call a new routing for the interconnection network. Finally, the method can reconfigure the network according to the new routing for the interconnection network.

What is claimed is: 
     
       1. A method for supporting efficient reconfiguration of an interconnection network having a pre-existing routing comprising:
 providing one or more computers, wherein each of the one or more computers comprises one or more microprocessors; 
 providing, at the one or more computers, a network, the network comprising a plurality of switches and a metabase, 
 storing, at the metabase, a first set of leaf-switch to leaf-switch multipaths, each of the first set of leaf-switch to leaf-switch multipaths being calculated by a management entity of the network; 
 upon detection, by the management entity, of a reconfiguration triggering event, distributing, from the metabase and by the management entity, at least one multipath of the first set of leaf-switch to leaf-switch multipaths to the plurality of switches; 
 updating, in parallel, a plurality of linear forwarding tables of the network based upon the distributed at least one multipath. 
 
     
     
       2. The method of  claim 1 ,
 wherein each of the plurality of linear forwarding tables is provided, respectively at a switch of the plurality of switches. 
 
     
     
       3. The method of  claim 2 ,
 wherein the plurality of switches comprises at least one physical switch and at least one virtual switch. 
 
     
     
       4. The method of  claim 1 ,
 wherein the reconfiguration triggering event is a performance-based triggering event. 
 
     
     
       5. The method of  claim 1 ,
 wherein the reconfiguration triggering event is a topology change of the network. 
 
     
     
       6. The method of  claim 5 ,
 wherein the topology change of the network comprises one of node failure and link failure. 
 
     
     
       7. The method of  claim 6 , further comprising:
 calculating, by the management entity, a second set of one or more leaf-switch to leaf-switch multipaths, the second set of one or more leaf-switch to leaf-switch multipaths being associated with the topology change of the network; and 
 storing, by the management entity, the calculated second set of leaf-switch to leaf-switch multipaths at the metabase. 
 
     
     
       8. A system for supporting efficient reconfiguration of an interconnection network having a pre-existing routing comprising:
 one or more computers, wherein each of the one or more computers comprises one or more microprocessors; 
 a network at the one or more computers, the network comprising a plurality of switches and a metabase, 
 wherein the metabase stores a first set of leaf-switch to leaf-switch multipaths, each of the first set of leaf-switch to leaf-switch multipath being calculated by a management entity of the network; 
 wherein upon detection, by the management entity, of a reconfiguration triggering event, at least one multipath of the first set of leaf-switch to leaf-switch multipaths is distributed, from the metabase and by the management entity, to the plurality of switches; 
 wherein a plurality of linear forwarding tables of the network are updated in parallel based upon the distributed at least one multipath. 
 
     
     
       9. The system of  claim 8 ,
 wherein each of the plurality of linear forwarding tables is provided, respectively at a switch of the plurality of switches. 
 
     
     
       10. The system of  claim 9 ,
 wherein the plurality of switches comprises at least one physical switch and at least one virtual switch. 
 
     
     
       11. The system of  claim 8 ,
 wherein the reconfiguration triggering event is a performance-based triggering event. 
 
     
     
       12. The system of  claim 8 ,
 wherein the reconfiguration triggering event is a topology change of the network. 
 
     
     
       13. The system of  claim 12 ,
 wherein the topology change of the network comprises one of node failure and link failure. 
 
     
     
       14. The system of  claim 13 , further comprising:
 wherein a second set of one or more leaf-switch to leaf-switch multipaths is calculated by the management entity, the second set of one or more leaf-switch to leaf-switch multipaths being associated with the topology change of the network; and 
 wherein the calculated second set of leaf-switch to leaf-switch multipaths is stored, by the management entity, at the metabase. 
 
     
     
       15. A non-transitory computer readable storage medium having instructions thereon for supporting efficient reconfiguration of an interconnection network having a pre-existing routing, which when read an executed cause a computer to perform steps comprising:
 providing one or more computers, wherein each of the one or more computers comprises one or more microprocessors; 
 providing, at the one or more computers, a network, the network comprising a plurality of switches and a metabase, 
 storing, at the metabase, a first set of leaf-switch to leaf-switch multipaths, each of the first set of leaf-switch to leaf-switch multipaths being calculated by a management entity of the network; 
 upon detection, by the management entity, of a reconfiguration triggering event, distributing, from the metabase and by the management entity, at least one multipath of the first set of leaf-switch to leaf-switch multipaths to the plurality of switches; 
 updating, in parallel, a plurality of linear forwarding tables of the network based upon the distributed at least one multipath. 
 
     
     
       16. The non-transitory computer readable storage medium of  claim 15 ,
 wherein each of the plurality of linear forwarding tables is provided, respectively at a switch of the plurality of switches. 
 
     
     
       17. The non-transitory computer readable storage medium of  claim 16 ,
 wherein the plurality of switches comprises at least one physical switch and at least one virtual switch. 
 
     
     
       18. The non-transitory computer readable storage medium of  claim 15 ,
 wherein the reconfiguration triggering event is a performance-based triggering event. 
 
     
     
       19. The non-transitory computer readable storage medium of  claim 15 ,
 wherein the reconfiguration triggering event is a topology change of the network. 
 
     
     
       20. The non-transitory computer readable storage medium of  claim 19 , wherein the topology change of the network comprises one of node failure and link failure, the steps further comprising:
 calculating, by the management entity, a second set of one or more leaf-switch to leaf-switch multipaths, the second set of one or more leaf-switch to leaf-switch multipaths being associated with the topology change of the network; and 
 storing, by the management entity, the calculated second set of leaf-switch to leaf-switch multipaths at the metabase.